Dopamine hydrochloride and 1,3,5-trimethylbenzene (TMB) were purchased from Aladdin Reagent (Shanghai, China). Ammonia aqueous solution (NH₃·H₂O, 30 wt%) was purchased from Macklin (Shanghai, China). Pluronic F127 was purchased from Sigma-Aldrich (MO, United States). Phosphate buffered saline (PBS) was bought from Gibco (Shanghai, China). Anhydrous ethanol was bought from Aladdin Reagent (Shanghai, China). Chloral hydrate was bought from Xiya (Chengdu, China). Dopamine hydrochloride was bought from Aladdin Reagent (Shanghai, China).

MPDA NPs were synthesized according to the procedure which had been previously reported (Chen et al., 2016). The structural and elemental distribution of MPDA NPs were detected by transmission electron microscope (TEM). The morphology of MPDA NPs was measured by scanning electron microscope (SEM). The particle size and zeta potential of MPDA NPs were determined by dynamic light scattering (DLS). The crystal structure was analyzed by X-ray powder diffraction (XRD). The surface area and pore diameter were determined by Brunauer–Emmett–Teller (BET).

Twelve BALB/c mice (female, aged 4 weeks old, weight 20 ± 2 g) were supplied by the Guangdong Medical Laboratory Animal Center (Guangdong, China). To get mice to acclimate to the new environment, they were fed in rearing rooms for a week before intervention. The temperature in the rearing room was controlled at 20 ± 3°C and the relative humidity was from 30 to 73%. The animals were housed (3/cage) and they were allowed to walk around the cage, eat and drink freely. The animal studies were approved by IACUC of The Eighth Affiliated Hospital, Sun Yat-sen University (2022-002-01).

After 1-week acclimation, 12 female BALB/c mice were randomly divided into 4 groups of 3 mice each (Table 1), and the mice of experimental groups were accepted MPDA NPs treatment for one time. In the following 7 days, the mice were allowed to walk around the cage, eat and drink freely, and their weight was recorded every day. On day 7, all mice were taken under anesthesia for cervical dislocation, and then organs were weighted and collected, including the heart, liver, spleen, kidney, and lung.

After the dissection of mice, the fat and fascia adhered to the exposed organs were carefully removed with surgical forceps, and then the organs were subsequently placed in cold PBS to wash away residual blood. After carefully drying, the organs were weighed immediately with an electronic balance. The following formula was used to calculate the organ indexes. o rgan   indexes ( % ) = ( W 1   ) / W 2 ∗ 100 % where W₁ is the weight of the organ and W₂ is the weight of the corresponding mouse.

After weighting, the needed tissues were immediately fixed in a 10% formalin solution. The fixed tissues were embedded in paraffin blocks and then sectioned into 5 mm sections and mounted on the slides. The slides were observed after hematoxylin and eosin staining (HE). Using a light microscope, the images were observed at a magnification of 100X.

The raw sequencing data obtained from the Illumina platform contained sequence artifacts, including reads containing adapter contamination, low-quality nucleotides, and unrecognizable nucleotide (N), which would cause errors in the following data analyses steps. Therefore, the downstream analysis was based on clean data which was transformed from raw data using Fastp software (Chen et al., 2018) according to the following standard: 1) Discarding a paired-end reads if either one end contains adapter contamination; 2) discarding a paired-end reads if more than 10% of bases are uncertain in either one end; 3) discarding a paired-end reads if the proportion of low quality (Phred quality <5) bases is over 50% in either one end. Next, the paired-end were mapped to the mouse genome [mm10, Genome Reference Consortium Mouse Build 38 (GCA_000001635.2)] using HISAT2 software (Kim et al., 2019). Counts for each gene were obtained using HTSeq software (Anders et al., 2015). Sequencing statistics for each sample were summarized in Supplementary Table S1.

The differentially expressed genes between each treated group and the control group (n = 3 per group) were determined by the DESeq2 R package (Love et al., 2014), which is a popular method for differential analysis of count data. The DEGs were determined based on two criteria: 1) |log2 (fold change) | > 1 and 2) adjust p value <0.05. FPKM (fragments per kb per million reads) were transformed from count data in R (https://www.r-project.org/). To visualize the overall distribution of the DEGs, the volcano plots based on count data and the heatmaps based on FPKM data were created in R.

To determine the functional annotation, GO enrichment (Harris et al., 2004), and KEGG pathway (Kanehisa and Goto, 2000), enrichment analyses were performed for the DEGs using the clusterProfiler R package (Yu et al., 2012). Moreover, three categories were included in enriched GO terms, which are biological processes (BP), cellular components (CC), and molecular functions (MF). The enriched GO terms were selected based on two standards: 1) adjust p value <0.05 and 2) the number of unique genes in each GO term was more than one. Similarly, the enriched KEGG pathways were selected based on two standards: 1) adjust p value <0.05 and 2) the number of unique genes in each KEGG pathway was more than one.

Both TEM and SEM observations displayed spherical morphology and mesoporous structure with a clear arrangement and uniform size distribution of MPDA NPs (Figure 1A, Supplementary Figures S1A, B, C). As shown in Figure 1B, the DLS measurements showed that the average hydrated particle size was roughly 223 nm, and the zeta potential was approximately -30 mV, indicating that the MPDA NPs surface had a negative charge. At the same time, the element mapping of MPDA NPs showed that C, N, and O elements were evenly distributed throughout the nanoparticles, with the C skeleton forming the core structure of MPDA NPs (Supplementary Figure S2), which was consistent with the XRD results. As shown in Figure 1C, the XRD spectrum of MPDA NPs exhibited a wide peak centered at approximately 2θ = 22°, and the corresponding crystal plane was (002), which verified that the main composition of MPDA was carbon element (Chen et al., 2016; Peng et al., 2019; Wang et al., 2019). The broad peak also indicated that MPDA NPs did not belong to crystalline materials (Ili Balqis et al., 2017). In addition, the MPDA NPs showed a BET surface area of 28.882 m² g⁻¹ (Figure 1D), which enabled them to load the drug efficiently. According to the BJH model, the pore diameter of MPDA NPs was about 1.75 nm and its pore volume was 0.311 cm³ g⁻¹ (Supplementary Figure S1D).

In intravenous and intramuscular injection groups, mice were administrated with MPDA NPs suspended in PBS at 40 and 8 mg for a single time, and in the intragastric administration group, mice were administrated with MPDA NPs suspension at 50 mg at two points in time, 6 hours had passed between them. It was noticeable that the maximum tolerated dose (MTD) of intragastric administration was higher than that of other exposure routes. During the 7-day period, no mouse deaths were observed. Furthermore, no abnormal behaviors including vocalizations, labored breathing, difficulties moving, hunching or unusual interactions with cage mates were observed as well.

As shown in Figure 2A, the weight of mice in each group was relatively stable, but the weight of mice in the intramuscular injection group saw a slight decrease in the first 3 days. Organ indexes analysis was performed on the hearts, liver, spleen, lung, and kidneys of mice, and it was found that organ indexes did not alter significantly after different exposure routes to MPDA NPs (Figures 2B–F).

MPDA NPs did not induce any changes in both the shape and volume of the organs on day 7 by different exposure routes, while the color of the spleen and liver in the intramuscular injection group and spleen, liver, and lung in the intravenous injection group became darker than the control group (Supplementary Figure S3). Furthermore, Figure 3 depicted the micromorphology of the organs of mice in each group after exposure to MPDA NPs. It could be observed that a small amount of MPDA NPs deposited in the liver, spleen, and lungs in the intravenous injection group, while no nanoparticles were observed in the organs of mice in the other two treated groups. It meant that some MPDA NPs could deposit in tissues passing through biological barriers and could not be excreted within 7 days. In addition, compared to the control group, the organs of mice in each experimental group did not display any histopathological changes.

The number of DEGs between the treated groups and control group were list in Table 2 and Supplementary Table S2. Furthermore, Figure 4 display the heatmaps and volcano plots of DEGs between the intragastric administration group and control group, for the five tissues, including heart, liver, spleen, lung and kidney separately. Also, the DEGs of intramuscular group and intravenous group were shown in Supplementary Figure S4 and Supplementary Figure S5. For the intragastric administration group, the organ with the largest number of DEGs was liver, up to 519. The rest of the organs had close numbers of DEGs, which were 37, 29, 52 and 22 identified in heart, spleen, lung and kidney separately. For the intramuscular injection group, the organ with the least number of DEGs was lung, down to 23. And 58, 64, 32 and 34 DEGs were identified in heart, liver, spleen, and kidney separately. For intravenous group, the organ with the least number of DEGs was lung, down to 11. And 124, 52, 153 and 93 DEGs were identified in heart, liver, spleen and kidney respectively. Furthermore, the number of up-regulated and down-regulated DEGs were close for each comparison (Supplementary Tables S3–S5).

To determine the function of DEGs and metabolic pathway enrichment, GO (Table 3) and KEGG pathway analyses (Table 4, Table 5) were performed on DEGs in each group. For intragastric administration group, we obtained 6 enriched GO terms, including regulation of cellular response to androgen receptor binding (GO:0050681), polymerase activity (GO:0003899, GO:0034062, GO:0097747), and GTPase binding (GO:0017016, GO:0031267) (Figure 5, Supplementary Table S6). Only one enriched KEGG pathway was observed, which is calcium signaling pathway (mmu04020). For intramuscular injection group, we obtained 39 enriched GO terms, including transmembrane transporter activity (GO:0008509, GO:0046943, GO:0051184, GO:0046873, GO:0072349, GO:0005342, GO:0022803, GO:0015081), channel activity (GO:0015267, GO:0022836, GO:0005217, GO:0005216, GO:0022839, GO:0099094, GO:0022834, GO:0015276, GO:0022838) and so on (Supplementary Figure S6, Supplementary Table S7). No enriched KEGG pathway was observed in this group. For intravenous injection group, we obtained 4 enriched GO terms, including catenin complex (GO:0016342), apical junction complex (GO:0043296), mismatch repair complex binding (GO:0032404), and catalytic activity, acting on RNA (GO:0140098) (Supplementary Figure S7, Supplementary Table S8). No enriched KEGG pathway was found in this group as well.